Hierarchical clustering based facial expression analysis from video sequence

Facial expression analysis from video sequences has been an active research area from the last two decades. There are plenty of algorithms for classification of six prototypic facial expressions from video sequences. In present paper we have proposed a novel method of hierarchical clustering for classification of facial expressions. The optical flow algorithm is used for detection of muscle movement arises due to different expressions. The maximally deformed facial regions (RoI) are extracted. Statistical features in these RoI are utilized for classification. The hierarchical Fuzzy C-Means algorithm classifies the six basic prototypic expression.